Search

A novel iris segmentation technique based on active contour is proposed in this paper. Our approach uses innovative algorithms, including two important ones, pupil segmentation and iris circle calculation. With our algorithms, we are able to find the center position and radius of pupil correctly and segment the iris precisely. The accuracy of our proposed method for ICE dataset is around 92% and also reached high accuracy level of 79% for UBIRIS. Our results demonstrate that the proposed iris segmentation method can perform well with high accuracy and better efficacy for Iris segmentation in images. Through a relatively high-performance algorithm to further cut up the round out the picture of the pupil conversion cutting growth square picture in order to make the judgment for biometric applications.

In this research, we propose a state-of-the-art 3D finger gesture tracking and recognition method. We use the depth sensors for both hands in real time music playing. In line with the development of 3D depth cameras, we implemented a set of 3D gesture-based instruments, such as Virtual Cello and Virtual Piano, which need precise finger tracking in 3D space. For hands tracking, model-based tracking for left hand and appearance-based tracking for right hand techniques are proposed. To detect finger gestures, our approaches consist number of systematic steps as reducing noise in depth map and geometrical processing for Virtual Cello. For Virtual Piano, we introduce the Neural Network (NN) method to detect special hand gestures. It has Multilayer Perceptron (MLP) structure with back propagation training. Literature has few examples using touch screen as medium, with fixed-coordinates, and 2D–gestures to control MIDI input. The end users should no longer carry anything on their hands. We use Senz3D and Leap Motion due to a few technical benefits. Senz3D and Leap Motion use a closer distance to hands, thus detailed finger gestures can be precisely identified. In the past years, we announced a set of virtual musical instruments and the MINE Virtual Band. Our research work is tested on lab environment and professional theatrical stage. More information and demonstrations of the proposed method can be accessed at: http://video.minelab.tw/DETS/VMIB/ .

Human Computer Interaction based Research has emerged in the early 1980s with the advent of computer technology. Human Motion Capture is the process of recording the movement of people. Among many kinds of human motion capture devises, Microsoft Kinect sensor and inertial sensors are most popular nowadays. In this paper we propose an efficient motion tracking mechanism to construct real time human skeleton animation using inertial sensors. We compare the results of our proposed method with the Microsoft Kinect sensor over the complicated motion tracking and joint position. During the experiment we observed that our results are much steady than Microsoft Kinect results. Some motions like hand cross over or leg cross over, our method showed better results than Kinect because the Kinect may lose skeleton of the blocked parts. On the other hand, since we use radio frequency inertial sensors, our method has a larger working area than Kinect.

3D model construction techniques using RGB-D information have been gaining a great attention of the researchers around the world in recent decades. The RGB-D sensor, Microsoft Kinect is widely used in many research fields, such as in computer vision, computer graphics, and human computer interaction, due to its capacity of providing color and depth information. This paper presents our research finding on calibrating information from several Kinects in order to construct a 3D model of a human object and to render texture captured from RGB camera. We used multiple Kinect sensors, which are interconnected in a network. High bit rate streams captured at each Kinect are first sent to a centralized PC for the processing. This even can be extended to a remote PC in the Internet. Main contributions of this work include calibration of the multiple Kinects, properly aligning point clouds generated from multiple Kinects, and generation of the 3D shape of the human objects. Experimental results demonstrate that the proposed method provides a better 3D model of the human object being captured.

An object surface extraction from a set of point clouds extracted using a Kinect V2 sensor is presented here. First several point clouds are extracted from different angles around the object. Second point clouds are registered with each other using the Iterative Closest Point algorithm. If the angle between two camera positions is large, point clouds are manually rough aligned. Then iterative closest point algorithm provides a finer alignment. Next the object boundary is found enclosing all the points in the aligned and merged point cloud. After that a triangular mesh is created with the boundary points. Finally colors acquired from the original point cloud are applied on the triangular surface generated. For a better visual effect colors of the points are interpolated on the triangular mesh generated.

Automatic text extraction from number plates is used in most of the countries in present. In this paper a semiautomatic text extraction from Sri Lankan vehicle number plates is presented. There are several number plate types currently available in Sri Lanka. First a set of template was created based on the font used in the number plates for all the English letters and numbers. Number plate first corrected as can be seen from a camera situated in front of the camera. Then the area of the numbers is extracted. Texts and numbers are then separated in to separate images. Finally separated texts and numbers are matched with the original template created using template matching. A data set consisted of 93 number plates is used to demonstrate the usability of the proposed method.

Panorama photos are very useful when capturing large scenic backgrounds like famous constructions or eye catching landscapes. Users normally prefer capturing such backgrounds with them or friends as the foreground, but the moving people in such popular locations always obstruct the iconic structure in the background. In this paper a solution for automatic panorama generation is presented which is capable of removing moving objects in the background. First user captures a short video starting from the centre of the focus area and then following a clockwise circular path. The proposed application generates the panorama based on input video following the steps: segmentation of human object, removal of moving human objects, and generation of panorama. Experimental output panorama photos show that the proposed system is very usable and results are satisfactory.